Collection databases, like the collections they record, are continuously changing entities shaped by human decisions and their cultural, economic and political contexts. Although analogous with collections, these databases have historically been used as management tools or for searching and sorting information related to objects. Over the past 20 years, recognition of the value of using collection data as a research tool for historical research, including provenance and the trade in cultural heritage, has grown. Through a comparative examination of projects that have used collection data within this context, this chapter will critically reflect on the ethical challenges and considerations of data integrity and data absence when working with museum collection databases. It will review projects based in the United Kingdom and Europe, that have engaged with diverse cultural heritage collections and their databases, timescales and perspectives, but share a common interest in looking across collections at scale. This chapter highlights the potential of large-scale analyses to augment what we can understand about museum collections and the wider art and antiquities market. Nevertheless, I argue that this potential is limited by the bias within, instability of and ultimately access to museums’ digital collection records.